Geographic Impact of Indian Economy Essay Example

a significant role in industrial ownership and production. The state's location considerations differ from those of the private sector. The private sector responds to strong state regulations, resulting in an industrial geography shaped by both economic geography and political economy factors.The text discusses the importance of analyzing the location decisions of firms in specific industries to understand the process of industrial location and concentration. Factors that may influence a firm's location decision include history, availability of infrastructure, proximity to buyers and suppliers, and local amenities, as well as local wages, taxes, subsidies, and incentives. The paper presents two exercises: firstly, the development and estimation of an economic model to assess the impact of region-specific characteristics on firm location choices in defined industries, and secondly, the study of industrial clustering within metropolitan areas. The empirical application in the first part uses establishment data for Indian industry to examine how regional characteristics contribute to location choices. The concept of regional characteristics extends beyond natural geography and focuses on economic geography, including the quality of transportation networks connecting the location to market centers, the presence of diverse buyers and suppliers that facilitate inter-industry transfers, and local amenities.Drawing on hypotheses from the New Economic Geography (NEG) literature, this analysis will provide a foundation for understanding if a region's economic geography affects firm-level location decisions. By explaining these decisions first, a framework can be built to evaluate the overall spatial distribution of economic activity and employment.

In the second part, we will investigate the industrial location process within metropolitan areas. Unlike the first part, we will not model intra-metropolitan industrial locations because there is not enough information about clustering in developing

nations to conduct hypothesis testing. Additionally, the necessary data is not available. Instead, we will take a different approach in two steps. First, we will examine patterns of industrial clustering and identify local clusters of specific industry groups. Then, we will attempt to explain these observed industry location patterns using various economic and political perspectives.

To analyze location choices, plant-level data from the Indian Annual Survey of Industries (ASI) for the years 1998-99 will be used. This data focuses on eight three-digit manufacturing industries: Food Processing, Textiles and Textile products (including wearing apparel), Leather and leather products.This text discusses the grouping of firms into specific sectors in order to analyze the impact of regional characteristics on different industries. Paper products, printing, and publishing, chemical, rubber, and plastic products, basic metals and metal products, and mechanical machinery and equipment are included in these sectors. By examining industries separately instead of as a whole, we can identify how regional factors affect them differently. For example, food processing is closely tied to rural areas, while machinery, metals, and electrical/electronics industries are more urban and benefit from agglomeration economies. The data used in this study include plant level data, district and urban demographic information from the 1991 Census of India, and detailed geographically referenced data on transport infrastructure. This information allows us to identify each plant at the district level spatially and at the four digit SIC level sectorally. The section of the paper discussing industry location at the national scale is divided into four parts.Firstly, we present the analytic framework and specify the econometric model for examining location decisions at the firm level. We then discuss the design of

spatial parameters and provide an overview of economic performance and spatial distribution in the selected industries. Moving on, we discuss the results from the econometric analysis. Finally, we conclude with a summary of the main findings and their implications for regional policy.

The analytic framework used to examine the location of manufacturing industry is primarily based on recent findings from the "new economic geography" (NEG) literature. Krugman (1991) and Fujita et al. (1999) have analytically modeled increasing returns, which arise from technological and pecuniary externalities.

In models of technological externalities, agglomeration is incentivized by interfirm information spillovers. Information is treated as a public good, and its diffusion generates benefits for each firm located in the region. The benefits of interaction increase with the number of firms, assuming each firm produces different information. As these interactions are informal, the extent of information exchange decreases with distance. This provides incentives for entrepreneurs to locate their firm in close proximity to other firms, leading to agglomeration.

Models by Fujita and Thisse (1996) and Fujita (1989) demonstrate that firms gain advantages from information spillovers within their local network. If the benefit to a firm at location x obtained from a firm at y is denoted as a(x, y), and f(y) represents the density of firms at each location y in X, then A(x) represents the aggregate benefit obtained by a firm at x from the available information in location X. Assuming that production utilizes land (Sf) and labor (Lf) with rents of R(x) and W(x) respectively at x, a firm located at x in X would maximize profits subject to the following equation: f * L(x) * W - S(x)

* R(x) + A(x) = P.

Furthermore, apart from the pure benefits of information sharing through proximity to firms in the same industry, there are also pecuniary advantages from sharing specialized input factors. The concentration of similar firms in a specific geographic area facilitates economies of scale in shared input production. Additionally, firms using similar technologies and facing common issues are more inclined to collaborate, sharing information on various matters such as problem-solving and development of new production methods. The benefits of locating near industry concentrations can be enhanced by the presence of interconnected industries. Research on industry clusters has greatly motivated the exploration of inter-industry externalities.

Clusters are geographically concentrated and interdependent networks of firms linked through buyer-supplier chains or shared factors (1) (2). The success of a cluster relies on the development of local linkages among firms, education and research institutions, and business associations (Porter 1990). This concept emphasizes interfirm relations that reduce production costs by lowering transaction costs among firms. Proximity of interrelated firms allows for reduced transportation costs for intermediate goods and easier sharing of valuable product information (Porter 1990). A well-developed network of suppliers in a region is an important factor for profit-maximizing firms when making location decisions. Additionally, transportation costs play a role in determining the location choice of firms, as manufacturing firms tend to locate in regions with larger market demand to realize scale economies and minimize transportation cost (Krugman 1991). If transport costs are very high, then activity becomes dispersed.In extreme cases, when a region is self-sufficient (autarky), each location must have its own industry to meet final demand. On the other hand, if transport costs are

negligible, firms can be randomly distributed since proximity to markets or suppliers does not matter. Agglomeration occurs when transport costs are at an intermediate level and labor mobility is low. We expect an inverted U-shaped relationship between spatial concentration and transport costs. To account for transport costs, we modify equation (2) to include the cost (TC) for the firm at location x. With lower transport costs, firms are encouraged to concentrate production in a few locations to reduce fixed costs.

By locating in areas with good access to input and output markets, firms can reduce transport costs. Being able to easily supply markets becomes a major driving force for agglomeration in locations where transport costs are low. Additionally, having high-quality infrastructure that links firms to urban market centers increases the likelihood of technology diffusion and knowledge spillovers between firms. It also enhances the potential for a diverse range of inputs.Analytical models in monopolistic competition illustrate that activities with increasing returns at the plant level tend to be attracted more to locations with good market access. The analytical framework in this section emphasizes the significance of economic geography in influencing both location and agglomeration at the firm level.

In accordance with insights from NEG (New Economic Geography) and regional science models, the concentration of industries, both individually and interconnected, the presence of reliable infrastructure to reduce transportation costs and improve market access, regional amenities, and economic diversity all play important roles in cost reduction, thereby affecting the location and agglomeration of industries. Next, we will discuss the economic geography variables used in this analysis. The following section will describe the econometric specification used to assess

the significance of these variables. The empirical strategy involves estimating a cost function to gauge how costs (and subsequently, profits) are impacted by the economic geography of the region in which the firm is situated. If specific factors related to local economic geography have a cost-reducing influence, firms are more likely to select regions with disproportionately higher levels of these factors. (3) Seven Economic Geography Variables Market Accessibility (MA) refers to access to markets, which depends on the distance from and size and density of market centers in close proximity to the firm.The classic gravity model, commonly used to analyze trade between regions and countries (Evennet and Keller 2002), states that the interaction between two places is proportional to their size, measured by population, employment, or other indicators, and inversely proportional to separation, such as distance. Building upon Hansen's work (1959), the formula for this model is a = j b ij j c i d S I, where Ii c represents the classical accessibility indicator for location i, Sj is the size indicator at destination j (e.g., population, purchasing power, or employment), dij measures the distance (or friction) between origin i and destination j, and b describes how distance affects interaction levels. However, empirical research suggests that the simple inverse distance weighting commonly used may not accurately reflect the real-world decline of interaction with increasing distance (Weibull, 1976).

A frequently used modified version of the model is a negative exponential one, shown as a - = j a d j ne i b ij e S I 2 2 /. In this form, Ii ne represents the potential accessibility indicator for location i based on

the negative exponential distance decay function. The other parameters retain their previous definitions, with the parameter a representing the distance to the inflection point of the negative exponential function.The development of accessibility indicators depends on the choice of distance variables used in the computation. In this analysis, we utilize network distance as the basis for the inverse weighting parameter, as outlined in Lall et al (2001). By employing information on the Indian road network system, as well as the location and population of urban centers (ML Infomap 1998), their accessibility index describes market access.

The concentration of firms within the same industry (localization economies) creates externalities that enhance the productivity of all firms in that industry (Henderson 1988, Henderson et al.1999, and Ciccone and Hall, 1995). There are multiple methods for measuring localization economies, such as own industry employment in the region, own industry establishments in the region, or an index of concentration reflecting a disproportionately high concentration of the industry in the region compared to the nation. To measure localization economies, we employ own industry employment in the district. This is calculated using employment statistics from the 1998-99 sampling frame of the ASI, which provides employment data for all industrial establishments in India.

The sample data utilized for estimating the cost function are drawn from this sampling frame.There are different ways to define the importance of inter-industry linkages in explaining firm level profitability and location decisions. These include input-output based, labor skill based, and technology flow based approaches. The most common approach is to use the national level input-output account to identify regional buyer-supplier

linkages. The presence or absence of these linkages at the local level can indicate the likelihood of a firm being located in that region. To evaluate the strength of buyer-supplier linkages for each industry sector, we calculate the sum of employment in a region weighted by the industry's input-output coefficient from the national input-output account.

To calculate the strength of the buyer-supplier linkage (LINK), we use the formula LINK = a = wj * ejr, where wj represents industry j's national input-output coefficient column vector and ejr represents total employment for industry j in district r. During the computation of this indicator, we observed that the industry categories in the NIC system and IO accounts do not have an exact match. Therefore, we created a concordance table to establish a connection between them before multiplying wj and ejr.Data on input output transactions are from the Input Output Transactions Table 1993-94, Ministry of Statistics and Programme Implementation. Economic Diversity: In addition to buyer supplier, there are other sources of inter-industry externalities. A prominent among these is the classic Chinitz-Jacobs' diversity. The diversity measure provides a summary measure of urbanization economies, which accrue across industry sectors and provide benefits to all firms in the agglomeration. Chinitz (1961) and Jacobs (1969) proposed that important knowledge transfers primarily occur across industries and the diversity of local industry mix is important for these externality benefits.

On the consumption side, the utility level of consumers is enhanced by increasing the range of local goods that are available. At the same time, on the production side, the output variety in the local economy can affect the level of output (Abdel 1988, Fujita

1988, Rivera Batiz 1988). In this study, we use the well-known Herfindahl measure to examine the degree of economic diversity in each district. The Herfindahl index of a region r (Hr) is the sum of squares of employment shares of all industries in region r. Unlike measures of specialization which focus on one industry, the diversity index considers the industry mix of the entire regional economy. The largest value for Hr is one when the entire regional economy is dominated by a single industry.The text explains that a higher value on the diversity index signifies a lower level of economic diversity. In order to interpret the measure more intuitively, the text states that Hr is subtracted from unity to calculate DVr, where DVr=1-Hr. It further explains that a higher value of DVr indicates that the regional economy is relatively more diversified.

In the next section, the text introduces the econometric specification used to test the effects of economic geography factors on the location of economic activity. It states that firms will choose a location if profits exceed a critical level demanded by entrepreneurs. The estimation methodology includes a cost function using micro-level factory data and economic geography variables, which can impact the cost structure of a production unit. Following the explanation of the methodology, a brief description of the data sources is provided.

Lastly, the text emphasizes that economic geography, or the characteristics of the region where a firm is located, is also an important factor influencing its cost structure.The production cost of a firm depends on its output, input value, transportation networks, input diversity, and technological externalities from similar firms in the region. These location-based advantages

influence a firm's location decision as they create cost-saving externalities. The modified cost function includes the influence of location-based externalities, where total cost is determined by firm-specific inputs, input prices, and location externalities. The model considers four conventional inputs - capital, labor, energy, and materials - and assumes four sources of agglomeration economies at the district level: market access, own industry concentration, buyer-supplier linkages, and regional diversity.The model's efficiency is significantly improved by a joint estimation of equation (4) and (5) with restriction (6). The final model includes two additional dummy variables to identify locational characteristics that might not be captured by agglomeration variables. These locations are categorized as rural, non-metro urban (D1), and metro urban (D2), with rural location serving as the reference category. We also use a dummy variable to test differences between public and private sector firms and examine if profitability varies by firm age. The impact of economic geography factors on the firm's cost structure can be evaluated by deriving the elasticity of costs with respect to economic geography variables. Equation (4) gives the cost elasticities as: Y A w A C ly q lq q j jl j l l ln ln ln g g g a + a + a + = (7). These location-specific externalities not only impact the cost structure but also influence factor demand. The impact of these variables on input demand can be derived from the cost share equations, where the cost share for input i, Si, can be expressed as wivi/C, with wi as factor price of input i, vi as the quantity demanded of input i, and C as total cost.The equation (8)

shows that i i i S w C v = and i i i w S C v ln ln ln ln - + =. Hence, the elasticities of input demands with respect to agglomeration factors Al can be expressed as l jl l l i A A C A v g + = ln ln (9).

To conduct our analysis, we utilize plant level data for 1998-99 obtained from the Annual Survey of Industries (ASI) conducted by the Central Statistical Office of the Government of India. The survey considers the "factory" or plant as the unit of observation and data is based on returns submitted by factories. Various firm level production parameters, such as output, sales, value added, labor cost, employees, capital, materials, and energy are utilized in our analysis (refer to Table 1.1 for more details).

In terms of factory level output, it is defined as the ex-factory value of products manufactured during the accounting year for sale. The measurement of capital is often carried out using perpetual inventory techniques. However, this approach necessitates tracking the sample plant over time, which poses challenges due to changes in sampling design and incomplete tracking of factories over time. In our study and within the ASI dataset, capital is defined as the gross value of plant and machinery. This encompasses not only the book value of installed plant and machinery but also the estimated value of rented-in plant and machinery. Doms (1992) has demonstrated that defining capital as a gross stock provides a reasonable approximation for capital measurement.Labor refers to the total number of employee man days worked and paid for by the factory in the accounting year.

The Indian ASI provides factory data for us to calculate input costs. Capital cost includes rent for land, building, plant, and machinery, repair and maintenance cost for fixed capital, and capital interest. Labor cost is the total wage paid to employees. Energy cost covers electricity, petrol, diesel, oil, and coal consumption. Self-generated electricity value is determined by the firm's average electricity purchase price. Material cost consists of the purchase value of domestic and foreign intermediate inputs. Capital price is the ratio of total rent to net fixed capital. Labor price is obtained by dividing total wage by the number of employees. Energy and material prices are weighted expenditure per unit output. Output value is weighted by factor cost shares. The ASI includes factories registered under sections 2m(i) and 2m(ii) of the Factories Act 1948 with 10 or more workers using power, and those with 20 or more workers not using power in the past 12 months.3 Goldar (1997) observes that factories are categorized by industries based on their main products, which can lead to reclassification of factories from one category to another in successive surveys, making it difficult to make inter-temporal comparisons. The quality of the data has been examined by cross-referencing with standard growth accounting principles and reviewing feedback from other researchers who have used the data. The geographical attributes allow us to identify each firm at the district level.

Spatial Distribution of Indian Industry
Before discussing the empirical analysis results, we provide a general overview of the concentration and key characteristics of firms in the study sectors.

First, we divide the economic landscape into non-urban areas, urban areas, and large metropolitan areas. The metropolitan

areas include Delhi, Mumbai, Kolkata, Chennai, Bangalore, and Ahmedabad, along with their urban agglomerations. Using sample data from the ASI for 1998-99, we observe that average wages across industries are highest in metropolitan areas (see Table 1.1). In comparison to the country's nationwide average annual wage of Rs.60,000 per employee, labor remuneration is Rs.74,000 for metropolitan areas, Rs.54,000 for other urban areas, and Rs.50,000 for non-urban areas. Among various industries, annual wages are highest in Electrical/Electronics (Rs. [remaining content is missing]).The wages in different industries vary significantly, with the highest average wage found in the IT industry at Rs. 101,000 per employee and the lowest average wage in the leather industry at Rs. 41,000 per employee. Additionally, even within sectors, wages tend to increase as we move up the urban scale.

Productivity indicators, such as output per employee and value added per employee, reveal interesting trends. Although several industries have high output per employee, the value added figures present a different scenario. For instance, in the computing and electronics industry, the per employee output is Rs. 344,000 but the value added per employee is only Rs. 65,000. Similarly, the output and value added per employee in the chemicals industry are Rs. 376,000 and Rs. 79,000 respectively, while in printing and publishing they are Rs. 314,000 and Rs. 204,000 respectively. This suggests that these industry sectors may not be efficient in converting inputs into higher value outputs.

In terms of spatial distribution, we use the Ellison Glaeser (1997) index of concentration to determine if industrial activity within sectors is concentrated in specific locations. The concentration index is defined as:

[1 - sum_{i=1}^{N} a_{i} left( frac{s_{i}}{M_{i}}
ight)^{2}]

where

r represents the level of geographic concentration of an industry, si is the region i's share of the study industry, xi is the regional share of total employment, H is the Herfindahl index measuring plant size distribution within the industry sector, and N represents the total number of regions.The EG index is based on the underlying principles of a firm's location choice and is influenced by the micro foundations. When plant locations are distributed randomly, the EG index approaches zero, highlighting the absence of a uniform distribution.